JP5202001B2 - Automatic cash transaction equipment management system - Google Patents

Description

  The present invention relates to a management system for an automatic cash transaction apparatus.

  Automatic teller machines are already widespread, and it is common for banks to install multiple automatic teller machines at one branch. When this number becomes widespread, the operational cost varies greatly depending on the management method of the transaction apparatus, so the importance of improving the efficiency of the management method is increasing. Patent Document 1 discloses a system that improves the efficiency of cash management of an automatic teller machine. The system uses a network to centrally manage the number of cash in a plurality of transaction devices in real time. As a result, it becomes possible to cope with the transaction apparatus before the cash becomes insufficient or full, thereby improving operational efficiency.

JP-A-8-320967

  For efficient operation of an automatic cash transaction apparatus, it is important to manage the state of dirt in the transaction apparatus in addition to cash. Since the transaction apparatus handles a large amount of cash at high speed, powdery dust generated by cash being scraped and oil stains attached to the cash are accumulated in the apparatus. Therefore, it is necessary to maintain the equipment regularly.

  When there are multiple transaction devices, maintenance is usually performed at the same time (for example, on the same day) for reasons such as work efficiency and labor costs. However, the frequency of use of multiple transaction devices installed is not the same. For example, if the transaction device near the entrance is compared with the farthest transaction device, the former is considered to be used frequently. If the frequency of use is high, it means that a lot of cash is put in, so a lot of dirt is accumulated. The maintenance timing needs to be determined according to the transaction equipment with a lot of dirt, but in that case, there is an inefficiency that maintenance is performed even for the transaction equipment with little dirt.

  The present invention has been made in consideration of the situation as described above, and in a cash automatic transaction system in which a plurality of cash automatic transaction apparatuses are connected to a host computer via a communication device, the plurality of cash automatic transaction apparatuses are , Information about each usage amount is transmitted to the host computer via the communication device. The host computer compares the usage of each automatic teller machine and displays an indication that it cannot be used for the automatic teller machine with a large amount of use. A message indicating that it is possible is displayed, and the customer is guided to an automatic teller machine with a small amount of usage.

  According to the present invention, the use frequency of a plurality of automatic teller machines is leveled, and the maintenance period can be extended.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  First, an overview of the system is explained. FIG. 1 shows a state in which a plurality of automatic teller machines 101 are connected to a host computer 100 via a network 120. The guidance device 102 installed in the automatic teller machine, which has a smaller usage amount than others, displays a display content 110 (hereinafter referred to as an available display) indicating that the device can be used. For example, a character string such as tradeable or available, or a specific color such as blue or green may be displayed. The other guidance devices 102 display the display content 111 (hereinafter referred to as “unusable display”) indicating that the device cannot be used. For example, a character string such as unavailable, unavailable, suspended, or in preparation, or a specific color such as red may be displayed. Thereby, it is possible to guide the customer to the automatic teller machine on which the available display 110 is displayed.

  As shown in FIG. 2, the host computer 100 includes a CPU 201, a main memory 202, an auxiliary storage 203 such as an HDD, and a communication device 204. The host computer 100 receives an identification error value that is an intermediate result of each banknote identification executed by each automatic teller machine 101 for each banknote and accumulates it in the auxiliary memory 203. This is called an identification error value table, and an example thereof is shown in a table 301 in FIG. The host computer 100 uses the values in the identification error value table 301 to estimate the amount of usage of each automatic teller machine.

  Next, the configuration of each device will be described, and the flow for receiving each value in the identification error value table will be described. After that, a specific method for estimating usage will be described.

  A block diagram of the automatic teller machine 101 is shown in FIG. Deposit / withdrawal unit 401, identification device 402 for identifying the denomination / true / false of cash, communication device 404 for connecting to a network, display device 405 such as an LCD, input device 406 such as a touch panel display, Guide device 104 that displays whether the device is available or unavailable, temporary stacker 410 that temporarily stores cash, transport device 411, reflux box 412 to 414 that can automatically store and take out cash, only automatic cash storage The automatic cash transaction apparatus 101 is composed of a non-reflux box 415 capable of performing the above operation and a control apparatus 403 for controlling each of the above apparatuses.

  An internal block diagram of the identification device 402 is shown in FIG. The identification device 402 includes a CPU 501, a main memory 502, an auxiliary storage 503 such as a nonvolatile memory, for example, sensors 504 to 506 such as an optical sensor and a magnetic sensor, and an IO 507 mainly for communicating with a control device.

  An internal block diagram of the control device 203 is shown in FIG. The control device 203 includes a CPU 601, a main memory 602, an auxiliary memory 603 such as an HDD, and an IO 604 for communicating with each device in the transaction device.

  A flow in which the automatic teller machine 101 acquires the identification error value and transmits it to the host computer will be described. The identification error value is acquired when performing a deposit transaction for depositing banknotes and a withdrawal transaction for depositing banknotes.

  First, the deposit transaction will be described with reference to FIG. First, the control device 403 controls the transport device 411 and transports banknotes from the deposit / withdrawal unit 401 to the identification device 402 (S701). Hereinafter, it will be described that the control device 403 controls the transport device 411 to transport bills, and that the control device 403 simply transports bills. The control apparatus 403 receives the identification result of each banknote identification with respect to one banknote (S702).

  Here, the identification processing of the identification device 402 will be described with reference to FIG. First, denomination identification that identifies whether a bill is, for example, 1,000 yen or 10,000 yen is executed (S801), and the identification result and the identification error value are stored in the main memory 502 (S802). Subsequently, authenticity determination for determining whether the bill is genuine or counterfeit is executed (S803), and the identification result and the identification error value are stored in the main memory 502 (S804).

  Subsequently, whether the banknote is clean enough to endure circulation or the banknote to be collected due to dirt or damage is executed (S805), and the identification result and the identification error value are stored in the main memory 502. (S806). The identification result of each identification is transmitted to the control device 403 (S807). If there are still banknotes to be processed, the process returns to step 801; otherwise, the process proceeds to step 809 (S808). In step 809, the identification error values of all processed banknotes stored in the main memory 502 are transmitted to the control device 403.

  The basic procedures of denomination identification, true / false identification, and damage identification executed by the identification device 402 are the same, and as shown in FIG. 9, first, a sensor signal is acquired (S901). Subsequently, feature extraction suitable for each discrimination is performed (S902), and the feature vector obtained as a result is applied to a discriminator to obtain a discrimination result (S903). Here, Mahalanobis distance, support vector machine, vector quantization, three-layer perceptron, etc. are used for the discriminator. Eventually, a threshold process determines whether the denomination is denomination or does not apply to any denomination. Hereinafter, when the denomination can be determined, the denomination is determined, and when not, the denomination is not determined. If it is true / false identification, it is judged whether it is a genuine note or not, and if it is correct / bad, it is judged whether it is a genuine note or not. Furthermore, generally, a value of an identification error can be obtained from the classifier.

  For example, when the Mahalanobis distance is used to determine whether a bill is a 1000 yen bill, the smaller the distance value, the more likely it is a 1000 yen bill. In this case, the distance value corresponds to the identification error value.

  In FIG. 7, the control apparatus 403 receives the identification result of each banknote identification with respect to one banknote (S702), and then determines whether or not the denomination is confirmed by denomination identification (S703). If it is confirmed, the process proceeds to step 704. Otherwise, the banknote is sent to the deposit / withdrawal unit 401 (S706). Next, it is determined whether or not the authenticity identification result is a genuine note (S704). If it is a genuine note, the bill is sent to the temporary stacker 410 (S705), otherwise it is sent to the deposit / withdrawal unit 401 (S706). It is determined whether there are still banknotes to be processed (S707). If there are still banknotes, the process returns to step 701, and if not, the process proceeds to step 708. In step 708, each identification error value for all bills traded from the identification device is received. Subsequently, the identification error value is transmitted to the host computer 100 (S709). The host computer receives the identification error value and stores it in the auxiliary storage 203.

  After the customer takes out the banknote sent to the deposit / withdrawal unit 401 in the process of step 706, the control device 403 starts processing the banknote sent to the temporary stacker 410 in step 705. This process will be described with reference to FIG. First, one banknote in the temporary stacker 410 is conveyed to the identification device 402 (S1001). The identification device 402 performs each identification in the flow shown in FIG. 8, and the control device 403 receives the identification result (S1002). In the denomination identification, it is determined whether or not the denomination is confirmed (S1003). If it is confirmed, the process proceeds to step 1004. Otherwise, the banknote is stored in the non-reflux box 415 (S1007). In step 1004, it is determined whether or not the authenticity identification result is a genuine note. If it is a genuine note, the process proceeds to 1005. Otherwise, the bill is stored in the non-reflux box 415 (S1007). In step 1005, it is determined whether or not the result of the damage identification is correct. If it is correct, the process proceeds to step 1006. Otherwise, the banknote is stored in the non-reflux box 415 (S1007). In step 1006, the control device 403 stores the banknotes in the reflux boxes 412 to 414 for each denomination according to the result of denomination identification in step 1003. Next, it is determined whether or not there is still a bill to be processed in the temporary stacker 410 (S1008). If there is still a bill, the process returns to step 1001, and if not, the process proceeds to step 1009. In step 1009, each identification error value in all traded banknotes is received from the identification device 402, and transmitted to the host computer 100 in step 1010. The host computer receives the above information and stores it in the auxiliary storage 203. This completes the processing related to the deposit transaction.

  In the flow of FIG. 7 and the flow of FIG. 10, the identification error value is received from the identification device and sent to the host computer in both flows, but it is executed only in either one of FIG. 7 or FIG. May be.

  Next, processing at the time of a withdrawal transaction will be described with reference to FIG. First, the control device 403 conveys one banknote to be withdrawn from any of the reflux boxes 412 to 414 to the identification device 402 (S1101). The identification device 402 performs each identification according to the flow of FIG. 8, and the control device 403 receives the identification result (S1102). In the denomination identification, it is determined whether or not the denomination is confirmed (S1103). If it is confirmed, the process proceeds to step 1104. Otherwise, the banknote is stored in the temporary stacker 410 (S1107). In step 1104, it is determined whether or not the authenticity identification result is a genuine note. If it is a genuine note, the process proceeds to step 1105. Otherwise, the bill is stored in the temporary stacker 410 (S1107). In step 1105, it is determined whether or not the result of the damage identification is correct. If it is correct, the process proceeds to step 1106. Otherwise, the banknote is stored in the temporary stacker 410 (S1107). In step 1106, the banknote is transported to the deposit / withdrawal unit for withdrawal. Subsequently, it is determined whether or not there are more banknotes to be withdrawn (S1108). If there are more banknotes, the process returns to step 1101, and if not, the process proceeds to step 1109. In step 1109, the banknote in the temporary stacker 410 is stored in the non-reflux box 415. Subsequently, an error value for each identification in all traded banknotes is received from the identification device 402 (S1110) and transmitted to the host computer 110 (S1111). This completes the processing related to the withdrawal transaction.

  A process for determining an automatic teller machine having a low usage amount in the host computer 100 and a process for guiding a customer thereafter will be described with reference to FIG. First, the usage priority of each automatic teller machine is calculated (S1201). The calculation method follows, for example, the following formula (1).

In equation (1), for the past Ni banknotes of denomination _i , the denomination identification, true / false identification, and damage identification identification error values are calculated, their weighted sum is taken, and the priority Pri is calculated. Calculated.

  Generally, when dirt is attached to the sensor, the identification error value increases. An example is shown in FIG. The horizontal axis represents time and the vertical axis represents the error value. As shown in the left diagram of FIG. 13, even if the identification error values in the plurality of identification devices are the same value at the start of operation, as shown in the right diagram of FIG. 13, a difference appears over time. . In this case, the identification device that gives the broken line identification error value is considered to be more conspicuous in the sensor. Equation (1) is an equation based on this concept, and has the effect of lowering the priority of a transaction apparatus having a large identification error value. In fact, it can be considered that the cause is other than dirt, but in any case, an increase in identification error indicates that identification is unstable and the possibility of misidentification increases. It is logical to lower the usage priority of a transaction device.

Next, a method for setting the constants a _{1, i} , a _{2, i} , a _{3, i} will be described here. When the distribution ratio mi of the denomination _i is obtained, the first setting method is possible and is set as shown in Equation (2). Distribution ratio m _i is a representation of relative distribution number for each denomination. For example, if the distribution ratio of 1,000 yen bills is 1,5,000 yen, the distribution ratio of 0.5 means that the circulation number of thousand yen bills is twice the circulation number of 5,000 yen. Since the distribution ratio m _i is meaningful in the relative ratio relationship of each value, even if all the distribution ratios m _i are multiplied by a constant, they have the same meaning.

If the distribution ratio _mi is unknown, all may be set to 1. In equation (2), b ₁ , b ₂ , and b ₃ are values indicating the weight of each identification. It is desirable to set the weights in order of increasing sensor signal values. If the authenticity identification uses more signal values from the sensor than the denomination identification and damage identification, for example, the set value of equation (3) may be used.

Alternatively, the setting value of Equation (4) may be used in a simplified manner.

In the second setting method, when most distribution intensive denomination was denomination i _0, reflecting only priority for the denomination i _0, using the condition of formula (5)

The constants a _{1, i0} , a _{2, i0} , a _{3, i0} are preferably set in the order in which the sensor signal values are used in large numbers. If true / false discrimination uses more sensor signal values than denomination discrimination and damage discrimination, for example, the set value of equation (6) may be used.

Alternatively, the setting value of Equation (7) may be used simply. In the example of Expression (7), the priority Pri of the identification device is calculated using only the identification error value for authenticity identification of the denomination i ₀ .

  After step 1201 of FIG. 12 is completed, information on whether or not each automatic teller machine 101 is currently trading is obtained through the network 120 (S1202). A message is transmitted to the automatic teller machine 101 that is not in transaction and has the highest priority Pri so as to display the available display 110 on the guidance display 102 (S1203). For other cash automatic transaction apparatuses 101 in non-transactions, a message is transmitted to display the unavailable display 111 on the guidance display 102 (S1204). This is the end of the processing for determining an automatic teller machine with a low usage amount in the host computer 100 and the processing for guiding the customer thereafter.

  Note that the flow shown in FIG. 12 may be executed for a limited time period. In times when users frequently visit the store, there is often no room for selection even if they choose the highest priority from non-transactions. Therefore, for example, the above flow may be executed only during a time period when the number of users is reduced, such as after 20:00 on weekdays.

The priority Pri may be calculated using linear regression as shown in Equation (8). In Eq. (8), first, linear regression of the value of the discriminant error value f _i (n) is performed for each denomination to obtain the coefficient u _{i, 0} and the constant term v _{i, 0} , and the priority Pri (n _p ) is Calculate using u _{i, 0} and v _{i, 0} .

Alternatively, quadratic regression may be used as in equation (9). In equation (9), first, quadratic regression of the value of the identification error value f _i (n) is performed for each denomination to obtain the coefficients u _{i, 0} , v _{i, 0} and the constant term w _{i, 0} , and the priority Pri (n _p ) is calculated using u _{i, 0} , v _{i, 0} and w _{i, 0} .

In Expressions (8) and (9), if n _p = 1, the current priority can be calculated. When n _p <1, this is the predicted value of priority. For example, if n _p = -9, it is the predicted value of priority when another 10 banknotes are received.

  In the first and second embodiments, values of denomination identification, true / false identification, and damage identification are used, but the sensor status may be grasped using a reference sheet. The reference sheet is stored in a reference box 1401 in FIG. The control device 403 passes the reference sheet through the identification device 401 (first timing) at a time outside the transaction of the automatic teller machine (time from the end of the transaction to the start of the transaction), and temporarily stores it in the temporary stacker 410. Then, it is stored again in the non-reflux box through the identification device 401 (second timing). The identification device 401 collects an image of the reference sheet at one or both of the first timing and the second timing, and compares it with the reference image read in advance from the auxiliary memory 503 to the main memory 502. For comparison, for example, as shown in Equation (10), the Euclidean distance between the two when all pixel values are regarded as vectors is used.

This comparison result is designated as priority Pri. The constants a _i may all be set to 1, for example, or may be set large for the sensor i important for identification. Further, it is not necessary to use all the pixel values, and the pixel values may be selected in a grid pattern. Alternatively, a large weight may be given to a pixel related to bill recognition accuracy, and a weighted Euclidean distance may be used.

  In Examples 1 and 2, the values of denomination identification, true / false identification, and damage identification were used, and in Example 3, the reference sheet was used. You may decide including. Expression (11) determines the priority Pri by the weighted sum of the number of banknotes traded, the number of deposit transactions, and the number of withdrawal transactions.

To consider only the number of banknotes traded, the number of deposit transactions, and the number of withdrawal transactions, a _{1, i} = a _{2, i} = a _{3, i} = 0 may be set. Examples of setting other weights a ₄ , a ₅ , and a ₆ are as follows. First, generally because transaction number of bills than the number of transactions is important, for example, if you set _{a 4 = 1, a 5 =} 0, a 6 = 0, will determine the priority Pri only deal number of bills . If the payment transaction number into account the withdrawal transaction number is set, for example, _{a 4 = 1, a 5 =} 0.5, and a ₆ = 0.25. In the deposit transaction, the banknote identification may be performed a plurality of times for one banknote. Therefore, it is preferable to make the weight of the deposit transaction more heavy than the weight of the withdrawal transaction. If a ₄ = 1, a ₅ = 1, and a ₆ = 1, the priority Pri is determined by giving the same weight to the number of bills to be traded, the number of deposit transactions, and the number of withdrawal transactions.

_{Incidentally, a 1, i, a 2} , i, by setting the value of a _{3, i} to a value greater than zero, considering the value of the identification error, the transaction number of bills, deposit transaction number, the value of the dispensing transaction number It is also possible to calculate the priorities. In this case, it is desirable to use the number of transaction banknotes and the number of transactions after maintenance.

  As mentioned above, although the Example regarding a banknote was shown, the same effect can be acquired even if it uses a coin. Moreover, you may use the identification error value obtained from the identification apparatus of both a banknote and a coin.

The figure which shows the structural example of a cash automatic transaction system. The block diagram of a host computer. The figure which shows the example of the identification error value table memorize | stored in the host computer. The block diagram of an automatic teller machine. The block diagram of an identification device. The block diagram of a control apparatus. The figure which shows the processing flow of the automatic teller machine at the time of payment. The figure which shows the processing flow of an identification device. The figure which shows the processing flow of identification. The figure which shows the processing flow of the automatic teller machine at the time of payment. The figure which shows the processing flow of the automatic teller machine at the time of withdrawal. The figure which shows the decision flow of the automatic transaction apparatus which carries out a guidance display. The figure which shows the time change of an identification error value. The block diagram of the automatic teller machine in Example 3. FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Host computer, 101 ... Automatic cash transaction apparatus, 102 ... Guide apparatus, 120 ... Network, 301 ... Identification error value table, 411 ... Conveyance apparatus

Claims (9)

Each is a cash deposit / withdrawal unit, a temporary stacker for temporarily storing cash, a transport unit for transporting cash, a reflux box for storing / removing cash, a non-reflux box for storing cash, and an identification device for identifying cash , A communication device, a display device, an input device that accepts input from a user, a plurality of automatic teller machines equipped with a control device that controls each device in the automatic transaction device,
An automatic cash transaction system having a host computer connected to the plurality of automatic cash transaction devices via a communication device,
The plurality of automatic teller machines transmit information on the respective usage amounts to the host computer via the communication device ,
The host computer receives information on each usage amount from the plurality of cash automatic transaction apparatuses, determines whether or not to start a transaction for each of the plurality of cash automatic transaction apparatuses, and uses the information as the plurality of cash automatic transaction apparatuses. Sent to the automated transaction equipment,
The information on the usage amount includes one or more identification error values calculated when the identification device in the automatic teller machine identifies cash denomination, authenticity, and harm,
The automatic cash transaction system, wherein the host computer calculates a use priority from the identification error value, and determines whether or not to start a transaction for each of the plurality of automatic cash transaction apparatuses . 2. The automated teller machine system according to claim 1, wherein the host computer receives information from each of the plurality of automated teller machines as to whether or not each of the devices is in transaction, and from among the automated teller machines that are not in transaction. An automatic cash transaction system characterized by determining an automatic cash transaction apparatus capable of starting a transaction. 3. The automatic cash transaction system according to claim 2, wherein the automatic teller machine receives information about whether or not its own transaction can be started from the host computer, and displays the information on whether or not the transaction can be started on the display device. Automatic cash transaction system. 4. The automatic cash transaction system according to claim 3, wherein the automatic cash transaction apparatus includes a second display device, and information on whether the transaction can be started is displayed on the second display device. . 5. The automatic teller machine system according to claim 3 or 4, wherein the automatic teller machine stores a medium serving as a reference in the reflux box, and the information on the usage amount is the reference collected by the identification device. An automatic cash transaction system characterized in that it is an error between a signal of a medium to become a standard signal determined in advance. 5. The automatic cash transaction system according to claim 3 or 4, wherein the information related to the usage amount includes one or more identification error values calculated when cash is identified by the identification device in the automatic cash transaction apparatus, and transaction banknotes. An automatic cash transaction system characterized by including any one or a combination of the number of sheets, the number of deposit transactions, the number of withdrawal transactions. 2. The automatic cash transaction system according to claim 1, wherein each of the host computers calculated when the denomination, authenticity, and correctness of one banknote is identified in the identification device in the automatic cash transaction apparatus. A weighted sum of one or more identification error values is calculated for a predetermined number of bills inserted in the past, and an automatic cash transaction apparatus capable of starting a transaction is determined using the sum of the weighted sum for the predetermined number of sheets. An automatic cash transaction system characterized by 2. The automatic cash transaction system according to claim 1, wherein each of the host computers calculated when the denomination, authenticity, and correctness of one banknote is identified in the identification device in the automatic cash transaction apparatus. A weighted sum of one or more identification error values is calculated for a predetermined number of banknotes inserted in the past, a regression analysis is performed on the weighted sum for the predetermined number of banknotes, and the weighted sum for the last inserted banknote An automatic cash transaction system characterized in that an automatic cash transaction apparatus that calculates an estimated value of the cash and determines an automatic cash transaction apparatus capable of starting a transaction using the estimated value. 2. The automatic cash transaction system according to claim 1, wherein each of the host computers calculated when the denomination, authenticity, and correctness of one banknote is identified in the identification device in the automatic cash transaction apparatus. When the weighted sum of one or more identification error values is calculated for a predetermined number of banknotes inserted in the past, the weighted sum for the predetermined number of sheets is regressed, and a predetermined number of banknotes are inserted in the future The automatic cash transaction system characterized by calculating the predicted value of the said weighted sum and determining the automatic cash transaction apparatus which can start a transaction using the said predicted value.

Images